118
B. M. J. K. Balasooriya et al.
and Rajasooriyar 2010; Kumar et al. 2016; Das et al. 2015). Elevated concentrations
of fluoride in groundwater are one of the well-explored contaminants inherent in Sri
Lanka (Villholth and Rajasooriyar 2010; Young et al. 2011; Dissanayake and Chandrajith 2007), while nitrate pollution is a global threat exerted on groundwater (Liu
et al. 2003; Villholth and Rajasooriyar 2010; Wongsasuluk et al. 2014). Groundwater
quality degradation may make it unusable for human use (Noshadi and Ghafourian
2016). Once a groundwater source is contaminated, its quality cannot be restored
back easily (Dohare et al. 2014). Hence, measuring both physical and chemical
parameters is important to estimate or predict the quality of groundwater (Liu et al.
2003). Therefore, groundwater quality assessment is a high priority concern under
current situation in Sri Lanka.
Mutivariate statistical analysis techniques are widely used to evaluate surface and
groundwater quality. Many studies have used cluster analysis (CA), factor analysis (FA), principal component analysis (PCA) and discriminant analysis (DA) for
data reduction and classification. These techniques enable us to obtain relationships
between parameters and sampling sites and also idea about the degree of similarities and dissimilarities among the various parameters present in sampling sites. This
identification can be extended to identify major factors and sources influencing the
groundwater quality, hence increasing the viability of groundwater quality monitoring and water resource management (Nosrati and Van Den Eeckhaut 2012; Noshadi
and Ghafourian 2016). Although there are several studies conducted on groundwater
quality in Sri Lanka, those studies have covered a selected area of the country, either
coastal aquifers or a selected area with specific health problems (Dissanayake and
Chandrajith 2007; Young et al. 2011; Chandrajith et al. 2014). In this background,
this study focused on conducting a groundwater quality assessment representing
groundwater wells located within entire country. Hence, water quality analysis was
conducted on a total of 1262 groundwater wells distributed all over the country for
18 water quality parameters. Thereafter, multivariate statistical analysis techniques
were applied to the data set. Cluster analysis was used to classify the groundwater
wells based on their quality. The most influential parameters on the classification
were identified through factor analysis and discriminant analysis. Using the evaluation results, finally it is aimed to increase the viability of water resource management
and protection in Sri Lanka by providing statistical analysis result as there is no such
study carried out for entire country.
5.2 Materials and Methods
5.2.1 Study Area
Sri Lanka is located in the Indian Ocean, close to southeastern coast of Indian subcontinent. Total area of Sri Lanka is 65, 610 km
2 and divided into 25 administrative
B. M. J. K. Balasooriya et al.
and Rajasooriyar 2010; Kumar et al. 2016; Das et al. 2015). Elevated concentrations
of fluoride in groundwater are one of the well-explored contaminants inherent in Sri
Lanka (Villholth and Rajasooriyar 2010; Young et al. 2011; Dissanayake and Chandrajith 2007), while nitrate pollution is a global threat exerted on groundwater (Liu
et al. 2003; Villholth and Rajasooriyar 2010; Wongsasuluk et al. 2014). Groundwater
quality degradation may make it unusable for human use (Noshadi and Ghafourian
2016). Once a groundwater source is contaminated, its quality cannot be restored
back easily (Dohare et al. 2014). Hence, measuring both physical and chemical
parameters is important to estimate or predict the quality of groundwater (Liu et al.
2003). Therefore, groundwater quality assessment is a high priority concern under
current situation in Sri Lanka.
Mutivariate statistical analysis techniques are widely used to evaluate surface and
groundwater quality. Many studies have used cluster analysis (CA), factor analysis (FA), principal component analysis (PCA) and discriminant analysis (DA) for
data reduction and classification. These techniques enable us to obtain relationships
between parameters and sampling sites and also idea about the degree of similarities and dissimilarities among the various parameters present in sampling sites. This
identification can be extended to identify major factors and sources influencing the
groundwater quality, hence increasing the viability of groundwater quality monitoring and water resource management (Nosrati and Van Den Eeckhaut 2012; Noshadi
and Ghafourian 2016). Although there are several studies conducted on groundwater
quality in Sri Lanka, those studies have covered a selected area of the country, either
coastal aquifers or a selected area with specific health problems (Dissanayake and
Chandrajith 2007; Young et al. 2011; Chandrajith et al. 2014). In this background,
this study focused on conducting a groundwater quality assessment representing
groundwater wells located within entire country. Hence, water quality analysis was
conducted on a total of 1262 groundwater wells distributed all over the country for
18 water quality parameters. Thereafter, multivariate statistical analysis techniques
were applied to the data set. Cluster analysis was used to classify the groundwater
wells based on their quality. The most influential parameters on the classification
were identified through factor analysis and discriminant analysis. Using the evaluation results, finally it is aimed to increase the viability of water resource management
and protection in Sri Lanka by providing statistical analysis result as there is no such
study carried out for entire country.
5.2 Materials and Methods
5.2.1 Study Area
Sri Lanka is located in the Indian Ocean, close to southeastern coast of Indian subcontinent. Total area of Sri Lanka is 65, 610 km
2 and divided into 25 administrative